Queenie
  1. Queenie
  2. Hits
  3. Retro photos
  4. GPT Restores Old Photos, and It Stunned
Retro photos · Xiaohongshu 3:4 carousel · potential 11 · Machine-translated

GPT Restores Old Photos, and It Stunned Me

Put the cover under the lamp. Each of the five things that made it work is a layer you can switch on: the words, the grid it sits on, the colours it uses, the subject, the light.

Cover of the original post: GPT Restores Old Photos, and It Stunned
Cover · 1 of 3Original on Xiaohongshu ↗
Layers
Spec sheet

1Hook

. The before-and-after difference in the same group photo is visible at a glance, making people stop to compare details and want to save it.

Text content
None; no readable text appears in the image
Information carrier
The difference between the same group photo in the top and bottom versions
Visual focal point
The yellowed damaged area in the top block is one focal point
1 likes 1 saves 1 shares

Posted by 叫我懒大王 on Xiaohongshu, 29 Apr 2026. Shown with credit and a link back.

Keep all five layers and swap what’s under them: your shop, your budget, your city. You get four covers and the post text.

The prompt, as used

Verbatim. Copy it, or let Queenie swap in your own details.

Using the reference case "GPT Restores Old Photos, and It Stunned Me" as a model for its old-photo restoration approach, period texture, visual storytelling, and posting tone, generate a new set of vintage old-photo image-text content. New topic: {fill in the photo topic you want to restore, rebuild, or recreate} Reference platform: Xiaohongshu Content direction: vintage old-photo image-text Creation method: keep the nostalgic emotion, film/photo-paper texture, character relationships, period feel of the scene, and restoration constraints of the original case; if it is photo restoration, it must stay faithful to the original image's identity, facial features, composition, and historical context, and avoid over-beautification, redrawing, false historical information, and a modern digital plastic look. Please output: 1. Cover title and subtitle; 2. A 6-9 page image-text script, with each page containing a visual description, main copy, period/texture keywords, and a save prompt; 3. Visual prompts that can be used directly for image generation or photo restoration, explaining requirements for people, scene, lighting, film grain, damage repair, and color restoration; 4. Posting suggestions: title length, tags, image ratio, and platform tone. Constraints: do not replicate the original author, original people, or brand logos; do not fabricate historical facts; when real people's photos are involved, prioritize identity consistency, and do not make them younger, beautify them, slim their face, or rewrite their facial features.

More retro photos recipes

All retro photos hits →
AI科技79
Retro photosXiaohongshu

AI Gives Old Photos New Life: Childhood Memories

12K likes 2.7K saves 760 shares
See the recipe
周一闭馆78
Retro photosXiaohongshu

The era when images and memory confirmed each other has ended

5.8K likes 1.1K saves 3.7K shares
See the recipe
料到Ai68
Retro photosXiaohongshu

How to restore pore-level detail from a blurry image?

2K likes 1K saves 191 shares
See the recipe
7.zip68
Retro photosXiaohongshu

AI video: A day in the life of an 80s "popsicle-selling woman."

9.6K likes 1.2K saves 61 shares
See the recipe
Icetea冻柠66
Retro photosXiaohongshu

Recreating the life of young people in 2008 with AI

1.1K likes 243 saves 476 shares
See the recipe
衣修哥66
Retro photosXiaohongshu

I'm Not Fixing Old Photos

6K likes 376 saves 111 shares
See the recipe